Implementation and longitudinal capture of routine patient-reported outcomes in non–small cell lung cancer care at a JCI-certified center in Colombia (2022–2025).

Abstract

e20105 Background: Routine patient-reported outcomes (PROs) support symptom monitoring and quality benchmarking, yet implementation data from Latin America are limited. We evaluated reach, longitudinal capture, attrition, and PRO score trajectories in a telephone-based program embedded within a non–small cell lung cancer (NSCLC) clinical care pathway at a Joint Commission International (JCI)–certified center since 2023. Methods: Retrospective cohort of eligible NSCLC patients treated at Fundación Santa Fe de Bogotá (July 2022–December 2025). PROs were scheduled at baseline, 6, and 12 months (predefined windows). Instruments included EORTC QLQ-C30 global quality of life (QoL) and dyspnea scales, GAD-7, and PHQ-9. Capture was a recorded score among due evaluations; non-capture was attributed to death or inability to contact after ≥6 attempts within the window. Data quality procedures ensured completeness of recorded scores and harmonized death status across instruments. Results: Of 105 eligible patients, 81 (77.1%) completed at least one PRO; 63.0% were female, 45.7% had stage III–IV disease, and 74.1% were treated with curative intent (ECOG missing 4.9%). Baseline capture ranged from 56.8% to 85.2% across instruments (Table). Capture declined over time, driven predominantly by inability to contact. At 6 months (due = 73), QoL/dyspnea capture was 39/73 (53.4%), with death 3/73 (4.1%) and inability to contact 31/73 (42.5%). At 12 months (due = 58), QoL/dyspnea capture was 28/58 (48.3%), with death 6/58 (10.3%) and inability to contact 24/58 (41.4%). After death harmonization, 12-month GAD-7/PHQ-9 capture was 18/58 (31.0%) with death 6/58 (10.3%) and inability to contact 34/58 (58.6%). Among paired assessments (n = 23), median QoL increased from 75 (IQR 62.5–83.3) at baseline to 83.3 (75–95.8) at 12 months (p = 0.015); median dyspnea remained 0. Median GAD-7 decreased from 2 (0–3) to 0.5 (0–1) and PHQ-9 from 2 (0–4) to 1 (0–1). Lower baseline QoL was associated with larger QoL improvement (p = 0.050), consistent with ceiling effects and survivorship bias. In exploratory models, systemic therapy was associated with lower 6-month inability to contact (odds ratio 0.25, 95% confidence interval 0.07–0.93). Conclusions: Telephone-based PRO collection achieved strong reach and baseline capture in routine NSCLC care, but sustained longitudinal capture was limited primarily by inability to contact. Instrument-specific denominators and harmonized death classification are essential for interpretable benchmarking; hybrid clinic-based and digital workflows may be required for durable longitudinal PRO monitoring in real-world settings.